Binge Eating Disorder (BED) is characterized by frequent and persistent episodes of binge eating and associated eating disorder psychopathology as well as general psychiatric comorbidity, psychosocial impairment, and obesity. Although a distinct clinical syndrome, BED is indistinguishable from bulimia nervosa on core eating disorder features such as eating, weight, and shape concerns. To date there is no definitive answer as to what constitutes the most efficacious treatment of BED. Specialty treatments (e.g., interpersonal psychotherapy [IPT]) have demonstrated significant short- and long-term reductions in binge eating and associated psychopathology, but not body weight. Behavioral weight loss (BWL) interventions produce clinically significant short-tem weight loss and reductions in binge eating among obese BED individuals, but the long-term impact on binge eating and other outcomes remains unclear. Guided self-help (GSH) has also demonstrated efficacy for BED in the short-term. The overall aim of this study is to compare the long-term effectiveness of three conceptually and procedurally distinct BED treatments: IPT, BWL, and GSH. A primary question posed is whether BED participants require a specialty treatment (i.e., IM, or whether BWL by itself is sufficient. BWL has the potential advantages of producing weight loss, at least in the short term and being more disseminable as a non-specialty treatment widely administered to obese individuals. Previous tests of BWL for BED have lacked adequate sample sizes, valid measurement of binge eating, and sufficient long-term follow-up. The proposed study remedies these methodological shortcomings, while also evaluating whether IPT or BWL have specific effects beyond that of a credible, minimal treatment (GSH) that is briefer, less costly, and more broadly disseminable than the professionally administered therapies of IPT and BWL. Further, the proposed study tests the differential effectiveness of these BED treatments across levels of negative affect, a dimension along which BED participants reliably subtype. Because IPT targets negative affect, we predict that IPT will prove more effective than either BWL or GSH for the subset of participants with high negative affect. 210 overweight BED participants will be recruited and divided into low negative affect and high negative affect subtypes. Participants will be stratified by negative affect sub-type and randomized to receive one of the three 24-week interventions: IPT, BWL, or GSH. Two treatment sites are required to recruit the large sample needed and to establish generalizability of findings. A third site will serve the function of data coordination and assessment/treatment oversight. Binge eating and associated outcomes will be assessed pre-treatment, post-treatment, and at 6-, 12-, and 24-month follow-up. In addition, likely mediators of change will be assessed during treatment. This study promises to not only clarify the most efficacious overall treatment for individuals with BED, but also to identify sub-groups for whom specific treatments are indicated.